by Claude Opus 5.5
What exactly is the norm “believe survivors”? An epistemic norm about credence, a practical norm about treatment, or a political slogan? Ferzan’s distinctions are a starting point.
“Believe survivors”, and its earlier variant “believe women”, has been invoked repeatedly in the public response to the case behind these seminars. It has been used to criticise the original decision not to charge, to defend the naming of the accused, and to rebuke those who urge caution. Critics reply that it is incompatible with the presumption of innocence. Much of the disagreement arises because the slogan is ambiguous. Kimberly Kessler Ferzan has argued that debates about #BelieveWomen conflate distinct questions: what to believe, what to do, who is being addressed, and in what context. Taking her distinctions as a starting point, this discussion examines three candidate interpretations and asks which, if any, survives scrutiny.
Reading one: an epistemic norm about credence
On the strongest epistemic reading, the norm says that one should believe, that is, assign high credence to, any allegation of sexual violence. This is untenable as a universal norm. It would require believing allegations that conflict with one another, and it would make credence insensitive to evidence, which is a defect in any epistemic norm.
A weaker epistemic reading is more defensible. It says that one’s prior credence in an allegation should be set appropriately, and that the appropriate prior is higher than the one historically applied. Rates of false allegations, though contested, are generally estimated to be low, and the history of systematic disbelief suggests that many people have applied priors depressed by stereotype. On this reading, “believe survivors” is a corrective: it tells hearers to adjust for a known bias and to refrain from demanding more corroboration of sexual offence complainants than of other witnesses.
This weaker reading is compatible with responsiveness to evidence. It does not say that an allegation must be believed whatever emerges; it says that disbelief should not be the default.
Reading two: a practical norm about treatment
On a practical reading, the norm concerns how to treat those who disclose: with respect, support and seriousness, without interrogation, blame or dismissal. This norm does not depend on any particular credence. One can treat a friend who discloses with care whether one’s credence is high or uncertain, and one should. Institutions can provide support and interim measures to a complainant without any finding about the truth of her allegations.
This reading is the most robust. It is addressed to everyone who receives a disclosure, including friends, teachers, university staff and police officers. Its rationale is not epistemic but moral: the harms of disbelief and hostility to a genuine victim are severe, while the costs of treating a false complainant with respect are comparatively small.
Reading three: a political slogan
On a political reading, “believe survivors” is not a norm of individual conduct at all but a collective demand for systemic change: an insistence that institutions stop dismissing sexual violence complaints, that the burden of disbelief be lifted, and that survivors be heard. Slogans function differently from norms; they compress complex positions into rallying cries, and their content is partly fixed by context.
As a slogan, “believe survivors” has been effective in shifting attention to under-reporting and institutional failure. But its compression creates risks. When it is read as an epistemic norm in its strong form, it collides with the presumption of innocence and with evidence-sensitivity. When it is applied to individual cases as a demand for a particular conclusion, it becomes a substitute for adjudication.
Ferzan’s distinctions applied
Ferzan’s key move is to separate law from life. The presumption of innocence governs what the state may do to an accused person; it does not dictate what private individuals must believe. Conversely, a private person’s belief in a complainant does not license state action against the accused. Once this distinction is made, much of the apparent conflict disappears: one can believe a friend and still insist that the state prove its case.
A second distinction concerns belief versus action. Belief can be appropriate where action is not. A person may believe, on the basis of a friend’s account, that an assault occurred, without being entitled to take actions against the accused, such as public denunciation, that require a higher warrant because of their consequences.
A third concerns roles. A fact-finder, such as a hearing panel or jury, has a role-specific duty to form beliefs only on the evidence presented and according to the applicable standard. For fact-finders, the strong epistemic norm is inapplicable; the corrective norm, against stereotyped disbelief, is essential.
Which reading should we adopt?
The defensible content of “believe survivors” is a combination: a practical norm of respectful and supportive treatment, addressed to all who receive disclosures; and a corrective epistemic norm against default disbelief and stereotype, addressed especially to institutional actors. The strong epistemic reading should be rejected. The political slogan is legitimate as a demand for institutional change but should not be applied to individual cases as a verdict.
Conclusion
In this case, the defensible version of the norm would condemn any early failure to take the complainant’s report seriously, any reliance on stereotypes in assessing her account, and any hostile treatment of her. It would not license public condemnation of the accused before the evidence is tested, nor demand a particular outcome from the special prosecutor. Understood this way, “believe survivors” and the presumption of innocence govern different domains, and each is stronger for being clearly distinguished from the other.